Georgia Institute of Technology
Computational Methods for Environmental and Social Good: Implications for Circular Waste Management
Abstract
dc:description.abstractIn the circular economy, a central point of discussion is the potential economic and environmental gains that emerge from more efficient waste management strategies and practices. This dissertation explores policy, infrastructural, and regional circularity aspects of the transition to circular waste management, focusing on plastic waste streams in the United States. It applies decision-support tools—including optimization, machine learning (ML), and artificial intelligence (AI)—to enable data-driven design of circular strategies and enhance policy evaluation related to the circular economy and climate regulations. The research is structured around three core applications in the context of circular waste management: (1) policy evaluation and inference, (2) circular infrastructure and strategic planning, and (3) analysis of regional drivers and barriers to circularity adoption. To begin, a generative AI approach for real-time evaluation of public perception toward circular waste policies is introduced. By analyzing social media discourse using Large Language Models (LLMs), this research assesses preferences regarding single-use plastics regulation and highlights LLMs as scalable, low-cost alternatives to traditional surveys. Building on this understanding of public engagement, the next focus turns to the practical implications for small and medium-sized enterprises operating within localized circular infrastructure. An optimization-based framework models a Polyethylene Terephthalate bottle manufacturer’s procurement strategy under extended producer responsibility and policy constraints, identifying trade-offs and decision-making strategies to mitigate resource scarcity, enhance supply chain resilience, and support circular business models. Finally, regional drivers and barriers to circularity adoption are investigated using interpretable ML to highlight key sociodemographic, infrastructural, environmental, and technological factors. A composite circularity index reveals critical influences on regional adoption of circular waste strategies in the U.S. By bridging computational methods with the systems-oriented perspective of Industrial Ecology, this research contributes to a growing body of literature on computational methods for environmental and social good. It provides scalable decision-support frameworks for researchers, policymakers and practitioners committed to accelerating the circular transition.
Degree
thesis:*- Level thesis:degree_level
- Doctoral
- Department dc:contributor.department
- Civil and Environmental Engineering
- Grantor dc:publisher
- Georgia Institute of Technology
- Year dc:date.issued
- 2025
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Hollauer, Catharina
- Advisors dc:contributor.advisor
-
- Bozeman, Joe F., III
- Asensio, Omar I.
- Committee members dc:contributor.committeemember
-
- Pecin, Diego
- Chen, Yongsheng
- Heidrich, Oliver
- Perez-Guzmán, Sofía
Subjects
dc:subject × 6Rights
- Language dc:language.iso
- en_US
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- https://hdl.handle.net/1853/82024